Stochastic Multi-objective Optimization For Integrated Energy System

Due to the fast depletion and severe pollution of fossil fuels,there is a massive stimulation to integrate the renewable energy resources(RES)such as wind and solar into power systems.With the growing penetration of RES in energy systems,it is imperative to assess and reduce the negative impacts caused by the uncertain RES.In the meantime,with the increasing proportion of the power generated by gas-fired generators in the total power production and the development of district energy demand,the interconnections among the gas network,the distributed district heating and cooling units(DHCs)and the electricity network become tighter,and their synergies would have a significant impact on the development of integrated energy system(IES).This thesis proposes a stochastic dispatch strategy based on the unscented transformation to tackle the uncertainty of RES.Furthermore,this thesis develops a many-objective optimization model for the coordinated optimal dispatch of an IES.The proposed model contains objectives representing the interests of the gas network,the distributed DHCs and the electricity network,as well as the environmental protection and system security.And an optimal formulation of the manyobjective optimization model is deduced in the thesis.At first,the development of power generation by the RES and natural gas in China is introduced.The modelling methods of uncertainty characterization for RES and the existing researches for the planning and operation of the integrated electricity and gas network are summarized subsequently.In addition,the techniques to tackle the many-objective optimization problems are comprehensively introduced,whereas the multi-attribute decision making methods are briefly introduced.The next,for the power systems integrated with wind and solar power,a sampling method named unscented transformation is suggested to characterize the uncertainty of wind and solar power considering the correlated relationship between them,and calculate the means and standard deviations(STDs)of system variables in accordance with uncertain RES.To reduce the cost and risk of a power system,the mean and STD of its generation cost are optimized coordinately,which is formulated as the mean-STD(MS)dispatch model.Then the MS model is solved by an algorithm named multi-objective group search optimizer with adaptive covariance and Lévy flights(MGSO-ACL),in which,a new constraint handling method is proposed to satisfy the operation limits of power systems under uncertainty.Finally,a decision making method named improved entropy weight is used to select a final optimal dispatch solution from the Pareto-optimal solution set output by the MGSO-ACL.The simulation studies and method comparisons are conducted on a modified IEEE 30-bus system to verify the feasibility and efficiency of the proposed method.Lastly,this thesis develops a many-objective optimization model,which contains objectives representing the interests of the electricity and gas networks,as well as the distributed DHCs,to coordinate the benefits of all parties participated in the IES.The objectives of the optimization model are the operation profit of gas network,the operation cost of distributed DHCs,the fuel cost,power loss,NOxemission,SO2 emission,voltage deviation,and voltage stability index of electricity network,respectively.In order to solve the many-objective optimization model efficiently,an objective reduction approach is proposed,aiming at acquiring the smallest set of objectives.The objective reduction approach utilizes the Spearman's rank correlation coefficient to measure the relationship between objectives based on the Pareto-optimal front captured by the MGSO-ACL algorithm,and adopts four strategies to reduce the number of objectives gradually.Simulation studies are conducted on a test IES consisting of a modified IEEE30-bus electricity network and a 15-node gas network.The results show that the many-objective optimization problem is transformed into a bi-objective formulation by the objective reduction approach.Furthermore,our approach improves the overall quality of dispatch solutions and alleviates the decision making burden to a great extent.